Virucidal activity by the copper-nickel-zinc alloy was tested in three main manners: wet virus droplet, dried virus droplet, and in solution with the alloy’s constituent metal ions. Each of these tests sought to elucidate different aspects in which the alloy surface causes inactivation and factors that may interfere in their testing. The wet virus droplet tests aimed to test virucidal activity using a current standard test, with a further investigation into the virus preparation methods that may interfere in those results. The dried virus droplet tests are more representative of hospital environmental conditions. Finally, the metal ion suspension tests sought insight into the virucidal activity of each major constituent metal for this particular alloy.
-
Wet Virus Droplet Tests
Virus droplet tests were performed at ≥ 90% relative humidity to prevent droplets from drying. These tests do not mimic realistic environmental conditions as virus will dry at the time-scale tested for the significantly lower relative humidity expected in hospitals. However, this method mirrors those outlined in a current standard tests evaluating surface disinfectants (JIS Z 2801, ISO 22196) and follows the methods described in Sect. 2.3.1.
-
Cell Debris as a Soiling Agent
To create a testable virus stock, virus was amplified by infecting a host cell culture and resulted in an approximate 10-fold increase in titer. When harvesting the amplified virus, cell viability was approximately 75% and density at 4 × 106 cells/mL. These parameters were selected to minimize cell debris and virus degradation from the proteins released by lysed cells. Low speed centrifugation was used to remove cells and large debris (i.e. clarification). Smaller cellular debris typically remains and therefore the supernatant was filtered using a 0.4 µm and then a 0.22 µm filter.
To determine the effect of cellular debris as a soiling agent, a representative amount of debris was selectively generated and added to the filtered virus stock. Debris generated during the amplification in a cell-culture based setting was selected. Based on culture conditions, debris generated by 1 × 106 lysed cells/mL is expected. Cells were grown to this density, collected using low speed centrifugation, resuspended in either fresh media or conditioned media, and sonicated to generate the simulated cellular debris soiling load. Conditioned media was taken from exponentially growing cells and low speed centrifugation was used for clarification. A further 1:10 dilution of the sonicated cells with the respective medias was also performed to generate a lower debris load. These solutions, along with fresh and conditioned media without debris, were used as diluents to investigate interference of the alloy’s virucidal effects by cell debris and growth media.
The tested conditions are outlined in the diagram in Fig. 1.
Results presented in Fig. 2A are measured after a 24 hour exposure to the copper alloy coated coupon. For conditioned media, addition of cells significantly prevented virucidal activity and was independent of the debris concentration (p < 0.001 for both). However, for fresh media, only the addition of 106 sonicated cells was significantly different from no cells (p < 0.05). Comparing the media for the same level of cell debris showed that the inactivation in conditioned media was significantly lower than that of fresh media.
-
Conditioned Media as a Soiling Agent
The protective effects of conditioned medium was investigated and is shown in Fig. 2B. Here, virus inactivation was investigated for up to 24 hours. Virus was suspended in either conditioned or fresh media using two independently prepared batches of working virus stock.
In Fig. 2B, points at or below the detection limit (denoted by an asterisk) were not included
in calculating the trend line. Model parameters were found to be significantly different between conditioned and fresh media (p < 0.05). These results indicate that virus in conditioned media followed a significantly slower inactivation (p < 0.05).
To determine if organic deposition from the soil was preventing ion leaching, ICP-OES was performed on these virus samples. These results are presented in Fig. 2C. Concentrations at each time point for copper, nickel, or zinc were not significantly different for soiling conditions (p < 0.05), indicating that soiling did not prevent leaching of metal ions.
-
Soiling Agents Used in Standard Tests
In developing a standard protocol for measuring antimicrobial activity of copper surfaces, the EPA initially proposed a soil made of PBS, FBS, and TX-100 (denoted Old EPA Soil). However, during public consultation of the protocol, it was suggested that the soil be changed to follow the current ASTM standard (denoted New EPA Soil) of FBS, yeast extract/yeastolate, and bovine/porcine mucin EPA [2016a]. Therefore, the effect of different soiling agents (conditioned media, Old EPA, and New EPA) was investigated.
No significant difference was found between any soiling condition at any time point (p > 0.05) as shown in Fig. 3. Solid lines connect the time points, while dashed lines indicate the next point was below the detection limit and are meant to help guide the reader’s eye. Points below the detection limit are indicated by an asterisk matching the sample colour.
-
Repeated Surface Exposures
This copper alloy surface is intended for a hospital setting where it is likely to become contaminated multiple times before being cleaned. A cursory investigation into the surface’s ability to repeatedly inactivate virus was performed. Further, a time series was used to discern any changes in inactivation kinetics.
Using three biological replicates, it was found that the surface was unable to repeatably in-
Figure 3: Titer of three different soiling conditions from 0 to 16 hours. Lines are not the result of any regression, but help to visualize trends in the data. Solid lines are between points in which all replicates were above the detection limit. Dashed lines indicate the next point had a replicate at or below the detection limit. Each point represents the geometric mean of two technical replicates and two biological replicates (n = 4). Error bars represent the standard deviation. Asterisks indicate at least one replicate was at or below the detection limit.
activate the virus. At 16 and 24 hours, the second exposure coupons had a significantly reduced inactivation of virus (p < 0.05), as shown in Fig. 4A. Between exposures, the surface was cleaned using 70% ethanol and gentle scrubbing under running deionized water. Due to the drastic decrease in virucidal activity between the first and second exposure, a third exposure was not performed. ICP-OES of the repeated exposure showed that copper and nickel leaching decreased significantly
(p < 0.05) on the second exposure (Fig. 4B).
Figure 4: A Copper alloy was subject to two rounds of virus exposure. The first exposure is presented in red points and lines, and the second exposure is presented in blue points and lines. Each point represents the geometric mean of three biological replicates (n = 3). Error bars represent the standard deviation. Asterisks indicate at least one replicate was at or below the detection limit. Lines are not the result of any regression, but help to visualize trends in the data. B ICP-OES of the samples collected for titering in Fig. 4A. Only the alloy’s major constituents are presented. Each bar represents the geometric mean of three biological replicates (n = 3), and are normalized to concentrations from the first exposure. Error bars represent the standard deviation.
-
Dried Virus Droplet Tests
In infected patients, expelled virus is typically in microliter volumes that dry fairly quickly in ambient humidities. The following tests evaluate the virucidal properties of the copper alloy surface using the methods described in Sect. 2.3.2.
-
Inactivation of Dried Virus
Inactivation of the dried virus was compared to stainless steel and the copper alloy. Drying of virus resulted in a 1-log loss (± 0.28), as shown by recovery from stainless steel (Fig. 5A). Virus was judged to have dried after 45 minutes in a biosafety cabinet using visual inspection. Copper alloy achieved inactivation beyond the detection limit (≥ 3.69-log) when in conditioned medium.
Virus stock, a control for all ambient factors excluding humidity, did not have a significant change (0.22-log ± 0.34).
-
Soiling Interference on Dried Virus Inactivation
Virus was suspended in fresh media, conditioned media, or new EPA soil. No significant difference
(p > 0.05) was found between conditioned media, fresh media, or New EPA soil and is presented in
Figure 5B. Virus in conditioned media and fresh media were inactivated beyond the detection limit (≥ 3.40-log ± 0.76 and ≥ 3.46-log ± 0.54, respectively), while the New EPA Soil samples still had detectable virus after drying (≥ 2.8-log ± 1.2).
-
Repeated Surface Exposures
Finally, the ability to repeatedly inactivate virus was tested. The same copper alloy surface was subjected to three virus exposures under the same soiling conditions. Between each exposure, the surface was cleaned by gentle scrubbing with 70% ethanol followed by a rinse with deionized water. Before reuse, the coupons were sterilized by immersing in 70% ethanol for at least 5 minutes.
Figure 5: A Virus was dried onto either copper alloy or stainless steel. Virus stock represents virus exposed to only the ambient temperature and light in a sealed micro-centrifuge tube. Each bar is the geometric mean of two technical replicates and two biological replicates (n = 4). Error bars represent the standard deviation. Asterisk indicates at least one replicate was at or below the detection limit. B Virus suspended in three different matrices was dried onto the copper alloy. Each bar is the geometric mean of two technical replicates and two biological replicates (n = 4). Error bars represent the standard deviation. Asterisks indicate at least one replicate was at or below the detection limit.
Figure 6: A As in Fig. 5B, virus suspended in three different matrices was dried onto the copper alloy. The alloy was exposed to each virus suspension a total of three times. Each bar is the geometric mean of two technical replicates and two biological replicates (n = 4). Error bars represent the standard deviation. Asterisks indicate at least one replicate was at or below the detection limit. B ICP-OES of samples collected in Fig. 6A. Only the copper ion concentrations are presented here for clarity. Error bars represent the standard deviation. Each colour corresponds to the exposure.
No significant difference was found between each exposure or soiling condition (p > 0.05 for both), as shown in Fig. 6A. These results indicate that dried virus was continually inactivated by the
surface.
For clarity, only the copper ions leached from the surface are presented in the ICP-OES data collected for these samples. The copper ion concentrations were also representative of alloy ions leached, as the ions leached in proportional concentrations.
-
Metal Ion Solutions
A factorial experiment was performed using divalent cationic copper, nickel, and zinc based on leachate concentrations from the alloy surface. The high ion concentrations were those determined from a leachate (Cu: 12.1 mM | Ni: 1.5 mM | Zn: 0.5 mM), the center-points concentrations were half of the high level, and the low concentrations were UPW only. The virus sample was prepared in two different ways, as described in Sect. 2.1 (filtered) or by omitting the filtration through a 0.2 µm filter (unfiltered). As determined in Fig. 3, the unfiltered virus sample contains a significant amount of cell debris which may be sequestering metal ions, resulting in a lower virus
inactivation.
Copper and nickel had a significant effect on reducing viral titer, and the interaction between copper and nickel was also significant for both methods of virus preparation as presented in Table 1 and Fig. 7. A significant difference was found between the preparation method for the centerpoint and combinations A, C, E, and G (Table 1. A further analysis on the unfiltered samples showed no significant difference between any of the copper combinations. Interestingly, the reduction by combination B (high zinc and nickel) was significantly greater than that of combination D (high nickel only). This difference indicates that there may be a synergistic effect between the two metals for virucidal activity.
Table 1
Combinations of divalent metal cations in a second factorial design of experiment (Fig. 7)
Combination | Copper | Zinc | Nickel | Log Difference |
Unfiltered | Filtered |
Center Point | Mid | Mid | Mid | -2.62 | -3.76 |
A | High | High | High | -3.57 | -4.72 |
B | Low | High | High | -1.23 | -1.24 |
C | High | Low | High | -3.26 | -4.79 |
D | Low | Low | High | -0.52 | -1.23 |
E | High | High | Low | -3.02 | -4.81 |
F | Low | High | Low | -0.04 | -0.01 |
G | High | Low | Low | -3.42 | -4.78 |
H | Low | Low | Low | 0.00 | -0.01 |